APPLICANT?S DESCRIPTION: This research concerns antisocial behavior (ASB), i.e., intentional behavior that harms or threatens other persons or their property. One of the more important empirical issues for this topic is identification of the risk factors in each age period that are associated with greater or lesser likelihood of subsequent ASB. In a previous meta-analysis, data were coded from 1,948 reports of findings from 251 longitudinal studies to generate a database of more than 15,500 correlational effect sizes representing the predictive relationships between risk variables and ASB across a wide age range. This project proposed here will, first, update and extend that database and, second, apply advanced meta-analysis techniques to synthesize the empirical research findings on four issues: 1. The capability of risk variables observed during different age periods to predict ASB at later ages. 2. The combinations of risk variables that co-occur during different age periods and the strength of their collective predictive ability when they are combined in multivariate factors. 3. The comorbidity of ASB and other problem behaviors at various age periods, e.g., the extent of co- occurrence with drug and alcohol use, school problems, mental illness. 4. Differences in predictive relationships between risk variables and ASB that are associated with age, gender, and ethnicity.